Andrew, I teach a two-course statistics sequence in sociology using R. Here are my responses to your four questions:
Q1. and Q2. The tough part is how to integrate teaching statistics (i.e., theory) with learning R (i.e., practice). I solved this dilemma by providing all R code from my lectures beforehand and by doing R demonstrations during lecture and by hosting a weekly one hour R lab. You must choose the right textbook. Most R textbooks assume either facility with programming schema or statistics. Usually, students I teach bring neither. I chose Verzani's "Using R for Introductory Statistics" and supplement it with advanced R textbooks, websites, journal articles, etc. You must decide whether you will (or expect students to) write functions vs. use functions included in the base installation. You need to decide what set of add-on packages students in your discipline should be familiar with, if any. I try to use functions from the base installation as much as possible and then introduce a few packages that are absolutely necessary. You must decide whether you want students to use a GUI (e.g., R Commander, SciViews-R) or type commands. You must decide whether you want them to use a command/text editor (e.g., TinnR, RWinEdt). I decided against both using a GUI and command/text editor--these are skills students can learn on their own. Q3. I use a combination of both approaches and actually encourage students to share R code that they have written. Sometimes we all learn about a useful function and maybe a way to write better R code. I find that the theory and practice have to be combined because some students have little interest in quantitative methods but have to take the sequence because it is required. Q4. I allow students to collaborate on graded assignments in small groups (n=3). These graded assignments must be turned in and completed as independent work but each student typically feels more or less comfortable with certain topics. And as a small group they can rely on each other for support and encouragement. Those students who just want to pass the courses sometimes appreciate statistics or R more because of the enthusiasm of another student. And students who get it learn to how explain what they know to someone who may be struggling. Cheers, Tony ------------------------------------------------------------------ Tony N. Brown, Ph.D. Associate Professor of Sociology Faculty Head of Hank Ingram House, The Commons Research Fellow, Vanderbilt Center for Nashville Studies Vanderbilt University (615) 322-7518 (615) 322-7505 fax [email protected] _______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-teaching
